INDEX

Explanations

references to specific individuals and their associated works or achievements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

with "and" as a suffix

np_acts-logits-general · gemini-2.0-flash

words ending in -and

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Hickey

-0.47

skog

-0.46

rxjs

-0.45

pédie

-0.42

 Pryor

-0.41

Kro

-0.40

 screw

-0.40

 Firth

-0.40

kro

-0.40

xue

-0.40

POSITIVE LOGITS

 Dand

0.88

rand

0.85

LAND

0.84

Land

0.84

hand

0.83

 LAND

0.82

 Land

0.82

RAND

0.82

 land

0.82

land

0.82

Activations Density 0.479%

references to specific individuals and their associated works or achievements

with "and" as a suffix

words ending in -and

No Comments

No Known Activations